Re: Equilibration simulation too slow.

From: Christopher Gillespie (gillescche_at_gmail.com)
Date: Sun Dec 07 2008 - 10:43:17 CST

Diego,

Does the pairlist distance need to be that large? Moving it to 17 (2
A greater than your cutoff) should help.

Chris

On Dec 7, 2008, at 10:59 AM, Diego Alejandro Vargas wrote:

> Hello,
> I am running an equilibration simulation for a protein (fascin) in a
> water box on a linux-amd64 cluster with Rocks OS. I had some trouble
> with it before, but I got to run with the help of people in the NAMD
> community, but to do so I had to increase parameters such as margin
> (10), cutoff (15), pairlistdist (25), and langevinPistonPeriod while
> reducing the timestep (0.5 fs).
>
> In this simulation I gradually raise the temperature by 10K every
> 500 steps until it reaches 310K, starting at 1K. In the first steps
> it starts very slow (~4s/step), but quickly levels off at ~0.7s/
> step. I know these exact numbers are irrelevant, but I have been
> told such a simulation should run faster.
>
> Does anybody have any suggestions on what other parameter could be
> altered to improve the efficiency? I am copying my configuration
> file below. Thank you for your attention.
>
> Diego A. Vargas
> ________________________________________________________________
>
> #############################################################
> ## ADJUSTABLE PARAMETERS ##
> #############################################################
> structure /wt_fascin_wb.psf
> coordinates /wt_fascin_wb.pdb
> bincoordinates /wt_fascin_wb_mn.restart.coor
> binvelocities /wt_fascin_wb_mn.restart.vel
> extendedSystem /wt_fascin_wb_mn.restart.xsc
> set temperature 310
> set outputname ***
> firsttimestep 0
> #############################################################
> ## SIMULATION PARAMETERS ##
> #############################################################
> # Input
> paraTypeCharmm on
> parameters ../common/par_all27_prot_lipid.inp
> # Force-Field Parameters
>
> exclude scaled1-4
> 1-4scaling 1.0
> cutoff 12.
>
> switching on
> switchdist 10.
> pairlistdist 13.5
> # Integrator Parameters
> timestep 0.5
>
> rigidBonds all
> nonbondedFreq 1
> fullElectFrequency 2
> stepspercycle 10
> margin 50.0
> # Constant Temperature Control
> langevin on ;# do langevin dynamics
> langevinDamping 5 ;# damping coefficient (gamma) of 5/ps
> langevinTemp 1
> langevinHydrogen off ;# don't couple langevin bath to hydrogens
> # Periodic Boundary Conditions
> # extendedSystem used.
> wrapAll on
> # PME (for full-system periodic electrostatics)
> PME yes
> PMEGridSizeX 100
> PMEGridSizeY 128
> PMEGridSizeZ 100
> # Constant Pressure Control (variable volume)
> useGroupPressure yes ;# needed for rigidBonds
> useFlexibleCell no
> useConstantArea no
> langevinPiston on
> langevinPistonTarget 1.01325 ;# in bar -> 1 atm
> langevinPistonPeriod 400.
> langevinPistonDecay 200.
> langevinPistonTemp 1
> # Output
> outputName $outputname
> restartfreq 500 ;# 500steps = every 1ps
> dcdfreq 250
> xstFreq 250
> outputEnergies 100
> outputPressure 100
> #############################################################
> ## EXECUTION SCRIPT ##
> #############################################################
> # Gradual temperature increase:
> run 500
> for {set i 10} {$i <= $temperature} {incr i 10} {
> langevinTemp $i
> langevinPistonTemp $i
> run 500
> }
> run 1000000 ;# 500ps

----------------------------------------------------------------------------------------
| Christopher Gillespie
| Ph.D. Candidate
| University of Delaware
| Department of Chemical Engineering
| Center for Molecular and Engineering Thermodynamics
| 150 Academy Street
| Newark, DE 19716
| gillesc (at) udel.edu
| http://udel.edu/~gillesc
----------------------------------------------------------------------------------------

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:13 CST